The present invention is directed to a brazing procedure for joining a ceramic or glass material (e.g., Al.sub.2 O.sub.3 or Macor) to graphite. In particular, the present invention is directed to a novel brazing procedure for the production of a brazed ceramic graphite product useful as a Faraday shield. The brazed ceramic graphite Faraday shield of the present invention may be used in Magnetic Fusion Devices (e.g., Princeton Large Torus Tokamak) or other high temperature resistant apparatus.
Radio frequency (RF) wave heating is a very attractive method for bringing plasma to ignition temperature in a magnetic fusion device from both the physics and technological standpoints. In a hot magnetized plasma, there are a wide range of energy absorption mechanisms of the electromagnetic wave offered by the plasma dielectric properties; there are several frequency ranges where plasma heating is possible. Each RF heating scheme selectively couples energy to a particular species of charged particles (ions and/or electrons) with the desired spatial deposition profile, to certain velocity distributions (Maxwellian or non-Maxwellian), and in either the perpendicular or parallel degree of freedom relative to the magnetic field.
Besides plasma heating, RF waves have many other potentially important uses in magnetic fusion devices, such as assisting initial discharge breakdown, driving a plasma current, controlling the temperature or current profiles, decreasing unwanted impurities, etc. To date, there are three major frequency ranges where plasmawave interaction are found to be strong, the ion cyclotron (ICRF), the lower hybrid (LHRF), and the electron cyclotron (ECRF) ranges. Both ICRF and LHRF ranges are below 10 GHz, so high power generator and transmission systems are available commercially. The ECRF range for future magnetic fusion devices, however, is between 100-200 where high power sources still need to be developed.
The advantages of RF systems in reactor environments are that wave launchers can be thermally shielded from the plasma and the RF sources can be placed behind radiation shields to avoid direct neutron streaming. This makes maintaining and operating the high-power system very convenient. Furthermore, since very efficient RF generators, transmission systems, and wave couplers exist over most of the RF range, the technological development needed is minimized. For RF heating schemes, the generator, transmission, and coupler systems still require technological studies; while the problem of wave absorption is in the primary area of interests of physics. In most instances, the engineering problems can be addressed independently of the physics questions by utilizing existing technology.
The wave launchers depend on the frequency range, the type of plasma wave, and the various plasma parameters. For the ICRF range, in experiments using present day machines, inductive coil antennas have been used because of low frequency (&lt;100 MHz) (i.e., the wavelength is much larger than the port dimensions on the devices.) As we approach magnetic fusion reactors, the dimensions of these devices become much larger, so it becomes feasible to use waveguide launchers for ICRF waves. The present LHRF heating experiments use frequencies between 800 MHz and 5 GHz where waveguide couplers can be employed, and the ECRF range is between 28 and 60 GHz where waveguides are used exclusively.
For the ICRF range, vacuum triode and tetrode tubes are available with power levels of 600-kW CW or 1.2-MW pulsed per tube. These tubes can be used in high-gain amplifiers with up to 85% efficiency (class D), and component costs in this frequency range are lower than other heating schemes.
In the LHFR regime, high-power klystrons are available with .about.50 kW per tube CW and 200 kW per tube pulsed, and the typical efficiency of these klystrons is between 50-60 percent. Another type of high-power source (.about.1 MW) that can be used in this frequency range is the gyrocon which has very high efficiency (.about.95 percent) but much lower gain (.sub..about..sup.&lt; 13 dB), so more amplifier stages will be required. Recent gyrocon development has extended the frequencies of these tubes to the 100-200 MHz range which is applicable to higher harmonic ion cyclotron heating.
Gyrotrons in the 28- to 60-GHz range with 100- to 200-kW pulsed or CW capability are presently used in tokamak and Elmo Bumpy Torus (EBT) experiments. 94GHz gyrotrons are being tested in the USSR on the T-10 tokamak with 200 kW of pulsed power. However, in reactors, the ECRF range is much higher (120 GHz&lt;f&lt;200 GHz) where high-power wave generators do not exist at this time. Therefore ECRF schemes, differing from the other two regimes, face a wave generator development requirement as well as plasma physics questions.
The present invention is directed to Faraday shields for magnetic fusion reactors which utilize ICRF principles. For a detailed discussion of these principles see Hwang et al, Radio Frequency Wave Applications in Magnetic Fusion Devices, Proc. of the IEEE, Vol. 69, No. 8, August 1981, pages 1030-1043.
Looking ahead to reactor applications of ICRF, the wave couplers, whether they are composed of waveguides, loops, or a hybrid configuration, must be designed to withstand the severe radiation and thermal environment. Design considerations can be classified into three areas: material selection, RF design, and installation convenience. The latter two points are more machine dependent since the port size and location determine the coupler shape and dimension; the materials problem, on the other hand, is more general. Recently at Princeton University, material selection and testing for near reactor requirements in radiation, thermal and RF power capability has begun in special testing facilities. Once suitable couplers are developed and tested, they can be placed on present day tokamaks to evaluate their wave generation properties.
Reactor material considerations can be divided into thermal properties and susceptibility to radiation damage. The projected thermal flux at the reactor first wall is 10-15 MW/M.sup.2, so some part of the RF coupler must act as a thermal shield to protect the rest of the structure. The estimated neutron fluence from a reactor is greater than 10.sup.16 N/cm.sup.2 /s; therefore, the materials used in the coupler must be able to withstand a total dosage of over 10.sup.23 N/cm.sup.2 in a three month period. In the couplers the integrity of mechanical supports, conductors, and insulators must be preserved within the necessary lifetime.
For the fast magnetosonic wave, it would be of great advantage if the Faraday shields could be used as the thermal shield for the conductors in loop couplers. Moreover, if the Faraday shield can be made vacuum tight, it could isolate the coupler from the plasma and local gas, thus minimizing RF breakdown problems in the launches. Accordingly, the development of a Faraday shield structure which would provide these properties would be a significant advance in the techonology.